Introduction

An enduring dispute in Late Pleistocene and Holocene archaeology of mainland Southeast Asia is the nature of the transition from forager economies to agricultural economies. As a key milestone in complex human-environment interactions, the debate has many dimensions. At one extreme of the debate is the claim that agricultural technologies and cultures appeared in Southeast Asia as a result of influence from north Asia, via the lower Yangtze River and the Yellow River. At the other extreme is the claim is that agriculture emerged from a locally contingent trajectory of changes in human-environment relationships (cf. Hunt and Rabett, 2014). While the cultivation of rice and the domestication of pigs and cattle took place in the Yangtze Valley earlier than elsewhere in mainland SEA (Chi and Hung, 2010; Higham et al., 2011), the influence of local contingencies remains poorly understood. One of the enduring challenges is that a critical period of time for this transition – the Late Pleistocene through to the middle Holocene – is sparsely represented in the archaeological record. We have a rich and well-documented record for the later Holocene when people were living more sedentary lifestyles, for example at Khok Phanom Di in Thailand and Man Bac in Vietnam. And we have many cave and rockshelter sites representing Pleistocene forager lifestyles, such as Tham Lod in Thailand and Xom Trai in Vietnam.

However, middle Holocene (c. 6000–3500 BP), the archaeological record in mainland SEA is particularly sparse. This major gap in archaeological evidence for the region has been called ‘the missing millennia’ (White and Bouasisengpaseuth, 2008:39). It is an important period because major changes occurred during this time. Ceramics appeared in many parts of Southeast Asia; domesticated plants such as millet and rice appeared; stone artefact technologies transitioned from mostly flaked to mostly ground stone artefacts; settlements expanded from primarily karstic upland and estuarine landscapes during the early Holocene to include inland alluvial lowland villages by the late Holocene (White, 2011). But the current sparsity of the archaeological record means that questions of the timing and character of these changes remain difficult to answer.

In this paper we present evidence of human activity from coastal Thailand that spans ‘the missing millennia’. Khao Toh Chong rockshelter is significant because it has a rich faunal record spanning the middle Holocene, and is located in an area with a well-documented sea level curve. This provides a unique opportunity to investigate locally contingent factors such as sea level changes on human subsistence behaviours at a critical time in the period of the transition from forager economies to agricultural economies. We report on a geoarchaeological analysis of the site to provide a local environmental context to the human occupation, as well as helping to understand site formation processes and artefact taphonomy.

Background

The Guangxi Province of southern China has extensive evidence of a forager economy with a semi-sedentary lifestyle during c. 7-4 k BP (Higham, 2013). Cave occupation continues until 6000 BC in Xianrendong and 5000–4000 BC in Zengpiyan, and more than 30 open sites containing shell middens have been found on the terraces of the Zuojiang, Youjiang and Yongjiang rivers near Nanning, in southern Guangxi (Chi and Hung, 2012; Fu; 2002). Occupation of these sites, characterized by the largest, Dingsishan, spans 7000-3500 BC. The sites include pottery manufacturing workshops, cemeteries and large quantities of aquatic and terrestrial animal bones, indicating that fishing and hunting were important activities (no cultivars have been recovered). The archaeology of this region gives the impression of a continuous sequence of human occupation. We see gradual, overlapping adaptations resulting in changes in landscape use, the appearance of pottery and use of cemeteries, and at a much later date, an agricultural economy. The pottery and burial practices of the Dingsishan shell middens is identical to those found at the Da But sites of northern Vietnam, such as Da But, Con Co Ngua, Ban Ban Thuy, Lang Cong and Go Trung. These sites were occupied by complex hunter-gatherer populations during 5500–2000 BC (Viet 2007). Polished axes, pestles and mortars suggest cultivation, but clear evidence of food production only appears around 1500-1800 BC at sites such as Man Bac with domesticated pig remains (Sawada et al., 2011).

While this gives a picture of continuity between complex hunter-gatherers and agriculturalists in southern China and parts of northern Vietnam, elsewhere in mainland Southeast Asia continuity is harder to see. Hang Boi cave in inland northern Vietnam has a thick shell midden that spans just 10,600-12,300 BP (Rabett et al., 2011). At sites in Thailand there is a gap between cave occupation and open site occupation. At Lang Rongrien rockshelter on peninsular Thailand, the most recent dated occupation is about 8000 BP, followed by undated and highly disturbed deposits containing burials and pottery (Anderson, 1990:20). Similarly, in northern Thailand rockshelter occupation at Tham Lod and Ban Rai becomes discontinuous at around 8000 BP (Marwick and Gagan, 2011; Shoocongdej, 2006). At Laang Spean rockshelter in Cambodia, the most recent occupation in 5000 BP, followed by later disturbance of the stratigraphy (Sophady et al. 2015; Forestier 2015). The general pattern seems to be that cave and rockshelter sites switch from being occasional habitation sites to burial sites in the middle Holocene (Anderson 1997). A key challenge here is that the burials disturb the stratigraphy, making it difficult to assess continuity between forager occupation and later activity. At open air sites, the record starts at around 2000 BC (i.e. 4000 BP), for example at Khok Phanom Di, near the Bang Pakong River, southeast of Bangkok (Higham and Thosarat, 2004) and at Ban Non Wat (Higham and Kijngam, 2011). Occupation at these sites is characterized by burials, pottery, and in later phases, polished stone artefacts indicating crop cultivation.

To investigate the gap in the archaeological record between the shift from rockshelters to open sites during the middle Holocene, we chose to focus on coastal karstic valleys of Krabi province. This landscape has been exposed to major changes as sea levels rose and fell during the Late Pleistocene and Early Holocene (Sinsakul, 1992). This makes it well-suited to assessing the effects of local environmental change on forager groups during a time of major transitions in subsistence. Sinsakul (1992) has documented a Holocene sea level curve for Thailand that starts with a steady rise in sea level until about 6000 BP, reaching a height of 4 m amsl (above mean sea level) (Figure 1). Sea levels then regressed until 4700 BP, then rising again to 2.5 m amsl at about 4000 BP. From 3700 to 2700 BP there was a regressive phase, with transgression starting again at 2700 BP to a maximum of 2 m amsl at 2500 BP. Regression continued from that time until the present sea levels were reached at 1500 BP. The evidence for these sea level changes comes from direct dating of marine shells and peat deposits at geological sites in peninsular Thailand (Sinsakul, 1992).

Figure 1: Holocene sea level curve for Thailand, redrawn from Sinsakul (1992). Data points are radiocarbon ages of geological specimens from beach and tidal locations.

Previous research into archaeological correlates of these sea level changes in peninsula Thailand have been summarized by Anderson (2005). He describes faunal evidence from Lang Rongrien that has increases in marine shellfish around 7500 BP and between 4000 and 2500 BP. Anderson proposes that the increases in marine shellfish at the site are probably related to increases in sea levels. A small number of other sites have been previously investigated in several provinces of peninsula Thailand. For example, Moh Khiew in Krabi with its 25,000 year old human remains (Auetrakulvit et al., 2012, Chitkament, 2007; Matsumara and Pookajorn, 2005; Pookajorn, 1994), Tham Khao Khi Chan in Surat Thani Province has occupation layers dating from 6060 BP to 4250 BP (Srisuchat and Srisuchat, 1992). Buang Bap, also in Surat Thani, has faunal remains including marine shellfish dating between 6000 and 5000 BP (Srisuchat and Srisuchat, 1992). Pak Om has a dense and diverse archaeological deposit, but its two dates of 9350 and 3010 BP come from the same layer, so the chronology is uncertain (Srisuchat, 1997). Khao Tau in Pang Nga is a site complex with deep stratification and abundant cultural materials dating to 4750 and 5250 BP (Srisuchat and Srisuchat, 1992), Finally, there is the Tham Sua shell midden in Krabi that is a deposit of marine shell greater than one meter deep and with a basal date of 6440 BP (Anderson, 2005).

These previous excavations demonstrate human occupation at several sites in peninsular Thailand during the critical time of sea level changes in the Holocene. However, the level of available detail does not give a clear picture of stratigraphic integrity or the subsistence strategies represented at the sites. The goal of our work at Khao Toh Chong was to build on this previous work, not only by collecting an assemblage spanning the Holocene, but also by conducting geoarchaeological analyses at the site to assess stratigraphic integrity and provide local environmental context to the human occupation.

Methods

Excavation methods

In June-July 2011 we excavated two areas of 2x2 m in 5 cm units to a depth of 1.6 m below the surface at Khao Toh Chong rockshelter. Excavated sediments were sieved using 5 mm and 10 mm screens. The site is a limestone overhang at the base of a 300 m high karst tower in Thap Prik Village. The rockshelter is about 30 m long with an average of about 10 m from the rear wall to the dripline. The dripline is about 40 m above the ground and a series of large boulders (3-4 m high) at the dripline give some protection from the wind and rain as well as trapping sediment in the shelter. The surface of the rockshelter is level fine sediment with no signs of disturbance and about 10 m above the surrounding ground, which is about 60 m above sea level. In Trench A, the southernmost trench, excavations reached a depth of 1.3 meters below the surface. In trench B, excavations were obstructed by bedrock in the northwest and southwest units. Subsequently, excavation depths in trench B extended to approximately 2.0 meters in the northeast and southeast units. Our archaeological and faunal analysis reported here is based on data from the southwest quadrant of trench A.

Geoarchaeological methods

Bulk sediment samples collected from a column taken from the south wall of excavation trench A. A 1 g sub-sample was dried at 60°C for 24 hours for particle size analysis. The sub-sample was sieved to remove the >2 mm particles and carbonates were removed by washing the sample in 20 ml of 1 M HCl. Samples were then centrifuged and treated with 30 ml of 30% H2O2 for an hour to remove organics (Scott-Jackson and Walkington, 2005). Additional drying occurred for 30 hours in a 60°C oven. Each sample was added to a mixture of deionized water and surfactant Triton X 10 and agitated before being run in a Horiba LA-950 at the University of Washington Materials Science Department. A quartz refraction index of 1.458 was used during analysis.

We measured pH and electrical conductivity (EC) using a portable Oakton Waterproof Dual Parameter PCSTestr 35 on subsamples with a 1:1 ratio of sediment to deionized water. Soil organic material (SOM) and calcium carbonate content were measured by the Loss on Ignition method (after Gale and Hoare, 1991), as the percent of mass lost after heating samples to 600ºC for 4 hours and 1000ºC for 2 hours. Magnetic susceptibility was measured using a Bartington MS2 Magnetic Susceptibility Meter with 10 cm3 of sediment analyzed in sample pots at low and high frequency. Three replicates for each sample measurement of low and high frequency susceptibility were taken following Gale and Hoare (1991).

Samples for organic Carbon isotope analysis consisted of a 2 g sub-sample dried at 60°C for 24 hours, sieved to remove the >2mm particle size fraction (Hartman, 2011), with organics were picked out and discarded before samples were ground for 5 minutes using a mortar and pestle. To remove mineral carbonates in the sample, 60 mL of 1 M HCl was stirred into the samples and left to sit for 24 hours while stirring every 10 hours (Millwood and Boutton, 1998). To rinse the HCl off the samples, 60 mL of deionized water was stirred into the samples for 1 minute before setting to dry at 60°C for 48 h. Two more rinse cycles occurred where 60 mL DI water was stirred into the samples for one minute before setting to dry at 60°C for 24 hours each. Isotope measurement was conducted using a Costech Elemental Analyzer, Conflo III, MAT253 at the UW Earth and Space Sciences IsoLab.

For XRD analysis, samples were scanned on a Bruker D8 Focus X-Ray Diffractometer with a Cu radiation source. Following McGrath et al. (2008) we sub-sampled 2 g of >2 mm sediment and ground it to a fine powder. Next 20 ml of 30% H2O2 was used to remove organic matter. After effervescence it was removed and dried for another 60°C for 24 hours. After a final grinding, samples were loaded onto trays and scanned on a Bruker D8 Focus X-ray Diffractometer from 5° to 75° 2θ with a Cu radiation source at resolution 0.02° steps per second with 40 kV and 40 mA power output. MDI Jade 9 software was used to identify minerals.

Samples were prepared for compositional analysis by ICP-AES as follows. A 1 g sub-sample was added to 10ml of HNO3 and heated at 90°C for 15 minutes (Misarti et al., 2011). Another 5ml of HNO3 was next added and heated at 90°C for 60 minutes to drive the reaction. Next, deionized water and 30% H2O2 were added for increased oxidization and 10ml HCl was added and heated for chloride formation. The samples were diluted with deionized water and filtered before being measured and placed into Falcon tubes for ICP-AES analysis. This acid digest provides a broad spectrum of elements in a known volumetric concentration, as required for ICP-AES analysis (Balcerzak, 2002; Carter, 1993). The samples were analyzed in a Perkin Elmer Optima 8300DV in the University of Washington Chemistry Department.

We were unable to extract quantifiable amounts of pollen or phytoliths from the sediment samples (further details are reported in Van Vlack, 2014). This is likely due to the frequent wetting and drying of the rockshelter deposit which creates poor conditions for microflora preservation.

Results

They key findings of our excavations were a faunal assemblage in a deposit with relatively few macroscopic traces of post-depositional disturbance (Figure 2). We did not encounter any burials or animal burrows and there is very limited termite activity in the deposit. We did not reach bedrock due to time constraints.

All excavated materials are currently stored at the Silpakorn University Faculty of Archaeology’s Phetchaburi campus. The raw data and code used to generate the results presented here have been organised into research compendium following the structure of an R package (Wickham, 2015) to enable reproducibility of the results (Marwick 2016). This compendium is archived online at DOI.

Figure 3: South section of Khao Toh Chong rockshelter trench A. The radiocarbon ages are the midpoints of the 95% calibrated age intervals. (c) indicates charcoal and (s) indicates shell as the material dated.

Figure 2: Plan of Khao Toh Chong rockshelter. The top image shows a view looking North, with trench A in the foreground. The middle image shows the South section of trench B. The bottom image shows the South section of trench A

Chronology

Sample code Age in years BP 1 sd error Material dated Excavation unit Context Depth below surface (m) Calibrated upper 95% Calibrated lower 95%
1 D-AMS 1140 149 25 charcoal 1 1 0.10 10.000 275.00
2 D-AMS 1141 178 26 charcoal 2 2 0.10 0.000 291.00
21 D-AMS 1142 1973 27 charcoal 4 3 0.20 1876.000 1985.00
3 D-AMS 1143 2846 30 charcoal 5 4 0.30 2878.975 3054.00
4 D-AMS 1151 5592 29 shell 6 5 0.40 6313.000 6424.00
5 D-AMS 1152 7051 50 shell 8 6 0.53 7765.000 7961.00
6 D-AMS 1146 11813 42 shell 13 7U 0.72 13558.000 13732.00
7 D-AMS 1149 11990 50 shell 15 8 0.95 13745.975 13980.00
8 D-AMS 1147 13026 45 shell 19 7L 1.15 15410.975 15736.02
9 D-AMS 1148 11236 42 charcoal 20 7L 1.25 13048.975 13168.00

Table 1: Summary of radiocarbon dates from Khao Toh Chong

Figure 4: Depth-age plot of calibrated radiocarbon dates from Khao Toh Chong

Charcoal and shell samples were dated using AMS methods by Direct AMS. Radiocarbon ages were calibrated to 95% ranges using Bchron 4.1.1 with the IntCal13 curve (Haslett and Parnell, 2008; Parnell et al., 2008; Reimer et al. 2011). Five charcoal samples and five shell samples returned radiocarbon age determinations. The shell’s ages are offset from the charcoal ages by an average of 3012 years, indicating a substantial reservoir effect. Considering only the charcoal dates, the excavated deposit spans about 13,500 cal. BP through to about 150 cal. BP.

The depth-age relationship for the dated samples is strongly linear, suggesting a constant rate of sediment accumulation. Although there is nearly a meter between the lowest and second lowest charcoal samples, the linear tendency of the shell samples that span this gap suggest that the accumulation of sediment at the site has been constant through the Holocene. Using the ages of the charcoal samples, we computed a simple linear regression model to estimate the approximate ages of undated excavation units. The regression equation is age = 11,088 * depth -666. Using this equation, we estimate the date of the lowest excavation level to be approximately 17,075 cal BP.

Geoarchaeology

Figure 5: Summary of bulk sediment analysis of samples from Khao Toh Chong

Analysis of sediments collected from the 2011 Khao Toh Chong excavations show a relatively constant depositional environment. The deposit is mostly sandy silt with occasional additions of coarser sands and gravels (for example in context A-4, 0.3 m below surface). Slight fluctuations in particle size distribution and carbonate percentage suggest minor variations in contributions from alluvial, fluvial and colluvial inputs (including limestone eroding from the karst tower). Overall the picture is of relatively constant and uninterrupted deposition.

Chemical analyses and magnetic susceptibility

The results of the basic chemical, magnetic susceptibility and particle size analyses are depicted in Figure 5. pH values at KTC are strongly alkaline throughout, with a shift occurring from pH 9.1 to 7.6 between contexts A-5 and A-6 (0.4-0.53 m below surface). Electrical conductivity (as a proxy for soluble minerals) and soil organic matter decline sharply below the surface, probably due to natural decay of organics. Soil carbonates are steady between 8% and 12% throughout. Low frequency magnetic susceptibility peaks at 308 in context A-5 (0.40 m below surface), indicating an enrichment of magnetic minerals in the deposit. This context also has the highest proportion of carbonates (12%), which would reduce magnetic susceptibility, so the change in A-5 is not a simple dilution of magnetic minerals by diagmatic minerals. Frequency dependency varies little, suggesting little change in the size of magnetic particles.

Carbon isotope analysis

The δ13C values range between -28.75 and -26.2, with values becoming increasingly negative in more recent times (Figure 5). This trend in δ13C values over time may be explained by isotopic fractionation and microbial activity (Lerch et al., 2011; Schweizer et al., 1999; Tieszen, 1991; Wynn, 2007), rather than major vegetation change throughout this period. This indicates an overall dominance of C3 plants, suggestive of forested-grassland vegetation, including evergreen trees and shrubs, surrounding the site (DeNiro, 1987; Yoneyama et al., 2010).

X-ray Diffraction

Context Quartz Calcite Kaolinite Periclase
B1 79.6 12.6 0.0 7.9
A2 66.1 11.1 19.9 2.9
A3 64.3 12.2 19.6 4.0
A4 89.5 7.8 0.0 2.7
A5 92.3 7.7 0.0 0.0
A6 68.9 9.5 19.0 2.6
A7U 80.5 19.5 0.0 0.0
A8 81.4 12.9 0.0 5.7
A7L 87.2 10.3 0.0 2.5

Table 2: Summary of X-ray diffraction data from Khao Toh Chong

The XRD analysis showed quartz and calcite present in all samples. Kaolinite was identified in samples from A-2, A-3, and A-6, suggesting a greater contribution of more intensely weathered sediment during the formation of those deposits (Alam et al., 2008). An alternative possibility is that the Kaolinite derives from ceramics found in those contexts. The proportions of Calcite in each sample support the loss on ignition results for carbonates, showing low variation throughout the sequence. Small amounts of Periclase were observed, indicating metamorphosis of the local limestone.

Inductively coupled plasma-atomic emission spectrometry

Ca Fe K Mg Mn Na Sr Ti Zn
KTC-B-1 5.68 5.19 4.29 5.01 4.24 4.45 2.30 2.89 2.94
KTC-A-2-DH 5.25 4.94 3.88 4.80 3.90 3.94 1.98 2.82 2.61
KTC-A-3-DH 5.17 4.85 3.80 4.70 3.81 3.71 1.94 2.71 2.49
KTC-A-4 5.64 5.18 4.17 5.00 4.14 3.91 2.40 3.14 2.85
KTC-A-5 5.48 5.04 3.90 4.86 3.95 3.61 2.21 2.87 2.68
KTC-A-6 5.46 5.10 3.88 4.91 4.01 3.49 2.19 2.77 2.77
KTC-A-7-up 5.61 5.26 4.23 5.09 4.15 3.56 2.47 3.18 2.93
KTC-A-8 5.44 5.15 3.99 4.77 4.05 3.27 2.27 2.96 2.68
KTC-A-7lo 5.57 5.23 4.14 4.86 4.13 3.35 2.49 3.07 2.82

Table 3: Elemental concentration by ICP-AES, all measurements are in ppm

Ca Fe K Mg Mn Na Sr Ti Zn
Ca 1.00 0.92 0.89 0.86 0.95 0.20 0.89 0.71 0.95
Fe 0.92 1.00 0.86 0.76 0.94 -0.09 0.97 0.82 0.91
K 0.89 0.86 1.00 0.82 0.95 0.35 0.82 0.76 0.92
Mg 0.86 0.76 0.82 1.00 0.81 0.34 0.71 0.66 0.93
Mn 0.95 0.94 0.95 0.81 1.00 0.24 0.86 0.71 0.96
Na 0.20 -0.09 0.35 0.34 0.24 1.00 -0.19 -0.16 0.25
Sr 0.89 0.97 0.82 0.71 0.86 -0.19 1.00 0.88 0.85
Ti 0.71 0.82 0.76 0.66 0.71 -0.16 0.88 1.00 0.70
Zn 0.95 0.91 0.92 0.93 0.96 0.25 0.85 0.70 1.00

Table 4: Correlation matrix of elements analysed by ICP-AES

Figure 6: Dendrogram of samples based on ICP-AES results

Results from ICP-AES analyses are presented in Table 3, with the concentrations of elements of interest to geogenic and anthropogenic sources including Si, Ca, Sr, Mn, Fe, Zn, Na, K, Mg, and Ti (Araujo et al. 2008; Arroyo-Kalin et al. 2009; Cook, 1965, Costa and Kern, 1999, Eidt, 1985, Knudson et al., 2004, Middleton, 2004, Middleton and Price, 1996, Woods, 1984 and Woods and Glaser, 2004). The majority of these elements are strongly positively correlated (Table 4), and there are no significant negative correlations. These relationships suggest a single source for the sediments throughout the entire period of deposition. Cluster analysis of the contexts using the elemental data suggests low-level groupings resulting from minor variation (Figure 6). The cluster containing B1, A4, A7U and A7L is notable because they are relatively enriched with Ca and Mg, but this is not correlated with carbonates measured by loss on ignition. Overall, the element distributions suggest low variation over time, and homogeneity in the composition of the deposit with a single source of sediment.

Archaeology

Figure 7: Examples of ceramics, ground and flaked stone artefacts from Khao Toh Chong. a) chert flake, b) quartzite flake, c) quartzite polished adze, d) chert flake, e) quartzite polished adze, f & g) ceramic sherd with incised and infilled decoration, h) cord-marked ceramic

Context Lithic count (n) Lithic mass (g) Ceramic count (n) Ceramic mass (g)
1 2 18.7 99 176.2
2 NA NA 1 0.9
3 NA NA 194 417.9
4 20 73.6 162 383.7
5 11 19.4 42 67.0
6 2 2.7 34 111.3
7H 2 6.3 24 38.1
8 3 3.5 NA NA
7L 8 57.8 2 42.2

Table 5: Summary of ceramics and stone artefacts recovered from Khao Toh Chong.

Figure 8: Distribution of ceramics and stone artefacts over time at Khao Toh Chong. Ages older than 13,000 cal BP have been extrapolated using the regression function described above.

The archaeological materials consist mostly of small broken pieces of ceramic and flaked stone artefacts (Table 5, Figure 7, Figure 8). Two complete polished adzes were found in the upper layers, and several flakes with traces of abrasion on the platforms were also found, indicating adze manufacturing. Ceramic decorations at KTC are typical for the region, including cord-marked and parallel incised and infilled lines. There are no significant correlations between the artefact counts and masses and any of the geoarchaeological variables. Overall, the assemblage is small and unremarkable. Artefacts were found in every excavation unit, but we suspect that ceramics in the lower part of the deposit may be post-depositional intrusions resulting from the activity of treeroots and termites. Similarly problematic associations of ceramics and radiocarbon-dated samples have been reported from Spirit Cave (Gorman ,1972). Radiocarbon dating of residues on ceramics at Spirit Cave obtained much younger dates (c. 3000 BP) than the stratigraphically associated charcoal samples (c. 7600 BP) (Lampert et al., 2003). We expect that future work on thermoluminescence dating of KTC ceramics may show a similar situation to that of Spirit Cave.

Despite this possibility of disturbance, we can make some brief observations from the archaeological sequence at KTC. The stone artefact technology changes from to large flaked cores and flakes made from coarse-grain metamorphic rock in the lower levels to polished adze flakes made from finer-grained rock in the upper levels. The ceramic assemblage also changes from thick, red sherds with frequent incised decorations in the lower levels to predominantly black sherds in the upper levels. However, the small number of artefacts in the deposit overall limits the degree to which we can distinguish these changes as part of a major regional trend or idiosyncratic use of this site.

Faunal assemblage

Taxon 1 2 3 4 5 6 7U 8 7L Total
Osteichthyes subclass indet. 0 (1) 5 (0) 0 (0) 0 (0) 0 (0) 0 (0) 2 (1) 0 (0) 0 (0) 7 (2)
Testudines fam. indet. 0 (1) 21 (1) 36 (1) 25 (1) 10 (1) 42 (2) 26 (1) 11 (1) 44 (1) 215 (10)
Varanus sp. 0 (1) 1 (0) 1 (1) 9 (1) 6 (1) 12 (1) 4 (1) 1 (1) 5 (1) 39 (8)
Pythonidae gen. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 1 (1)
Primates fam. indet. 0 (1) 2 (0) 1 (1) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0) 2 (1) 6 (4)
Macaca sp. 0 (0) 0 (0) 5 (1) 1 (0) 0 (0) 0 (0) 8 (1) 0 (1) 3 (1) 17 (4)
Trachypithecus obscurus 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (1) 2 (0) 0 (0) 2 (1)
Rodentia fam. indet. 0 (0) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 2 (1) 1 (0) 0 (0) 4 (2)
Rattus remotus 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (1) 1 (0) 0 (0) 0 (0) 1 (1)
Cannomys badius 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 2 (1) 2 (1)
Atherurus macrourus 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 1 (1)
Carnivora fam. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0) 1 (1)
Tragulidae gen. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (1) 1 (0) 0 (0) 1 (1)
Cervus unicolor 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (1) 1 (0) 0 (0) 1 (1)
Muntiacus muntjak 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 3 (1) 4 (2)
Bovidae gen. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 1 (1)
Total 0 (4) 29 (1) 44 (5) 35 (2) 16 (2) 56 (6) 44 (9) 18 (4) 61 (8) 303 (41)

Table 6: NISP of mammal, reptile and fish remains recovered from Khao Toh Chong (MNI values in parentheses).

Taxon 1 2 3 4 5 6 7U 8 7L Total
Amblemidae gen. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 56 (0) 72 (0) 143 (0) 372 (0) 1 (1)
Amphidromus atricallosus 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 1 (1) 1 (1) 0 (0) 0 (0) 2 (2)
Ampullariidae gen. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 3 (3) 0 (0) 3 (3) 7 (7)
Anadara sp. 0 (0) 0 (0) 1 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1)
Arcidae gen. indet. 0 (0) 0 (0) 0 (0) 3 (0) 0 (0) 1 (0) 0 (0) 0 (0) 0 (0) 11 (11)
Corbiculidae gen. indet. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0) 0 (0) 0 (0) 71 (71)
Cyclophoridae gen. indet. 0 (0) 0 (0) 2 (2) 3 (3) 2 (2) 1 (1) 28 (28) 5 (5) 30 (30) 27 (27)
Cyclophorus cf. saturnus 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (1) 11 (11)
Cyclophorus malayanus 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 9 (9) 1 (1) 0 (0) 1 (1) 2 (2)
Cyclophorus sp. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 7 (7) 0 (0) 0 (0) 1 (1)
Filopaludina sp. 0 (0) 0 (0) 0 (0) 2 (2) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 6 (6)
Muricidae gen. indet. 0 (0) 0 (0) 1 (1) 4 (4) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 6 (6)
Neoradina prasongi 0 (0) 0 (0) 8 (8) 134 (134) 52 (52) 82 (82) 1584 (1584) 2215 (2215) 771 (771) 4846 (4846)
Nerita balteata 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 2 (2) 0 (0) 0 (0) 0 (0) 14 (14)
Neritidae gen. indet. 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 6 (6)
Pila sp. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 2 (2) 4 (4) 0 (0) 12 (12)
Plectopylis degerbolae 0 (0) 0 (0) 0 (0) 1 (1) 0 (0) 1 (1) 2 (2) 3 (3) 5 (5) 3 (3)
Pseudodon sp. 0 (0) 0 (0) 0 (0) 1 (0) 9 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0)
Rhiostoma jalorensis 0 (0) 0 (0) 0 (0) 2 (2) 2 (2) 10 (10) 9 (9) 2 (2) 2 (2) 4 (0)
Rhiostoma sp. 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 6 (6) 3 (3) 2 (2) 10 (0)
Telescopium telescopium 0 (0) 0 (0) 0 (0) 3 (3) 2 (2) 3 (3) 4 (4) 2 (2) 0 (0) 643 (0)
Viviparidae gen. indet. 0 (0) 0 (0) 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0) 1 (0)
Total 0 (0) 0 (0) 13 (12) 155 (151) 67 (58) 167 (110) 1720 (1647) 2377 (2234) 1187 (815) 5686 (5027)

Table 7: NISP of mollusk remains recovered from Khao Toh Chong (MNI values in parentheses).

Context NTAXA Simpson Pielou
1 4 0.750 1.000
2 1 0.000 NaN
3 9 0.740 0.807
4 11 0.231 0.268
5 6 0.245 0.335
6 14 0.485 0.461
7U 20 0.085 0.092
8 11 0.020 0.033
7L 16 0.121 0.124

Table 8: Ecological indices of diversity and evenness for the faunal assemblage recovered from Khao Toh Chong. Pielou’s index is also known as the Shannon index of evenness

Mammalian abundance and distribution at the rockshelter throughout the late-Pleistocene and Holocene describes a diverse array of taxa in the deposits (Table 6). Although the majority of identified mammalian taxa represent a small sample size, there are several important patterns in the KTC assemblage. For example, the identification of large-sized artiodactyl taxa, including the Sambar deer (Cervus unicolor) and Muntjak deer (Muntiacus muntjak) at the late-Pleistocene and early Holocene period suggests that a more open and drier forest habitat surrounded the rockshelter during that time (Francis, 2008).

The values for species richness per context, of the mammalian, reptilian, and piscean taxa appears to be driven primarily by the presence or absence of carapace elements belonging to the Order Testudines, likely representing species of the turtle Family Trionychidae and Emydidae, based upon comparable faunal analyses at Lang Rongrien Rockshelter (Mudar and Anderson, 2007). Identification of abundant Varanus sp., and a moderate representation of Macaca sp., occurred in abundance with Testundines elements. Overall the presence of vertebrate remains, in comparison with invertebrate remains, was low at the rockshelter. Artiodactyl remains are notably restricted to the terminal Pleistocene and early Holocene deposits.

Of the identified mollusk remains, nine taxa were identified to the species level while an additional fourteen were identified to a broader degree of taxonomy (Table 7). Mollusk species richness varies between 0-11 species throughout the unit with a mean of 4.21 per context. Neoradina prasongi shells are of the most abundant species in the assemblage, specifically during the late-Pleistocene and early Holocene. When combined with shells from the Family Amblemidae and Cyclophoridae, these three taxa account for 97% of the identified mollusk taxa at KTC.

For all identified fauna, MNI and log NISP values for each context are strongly correlated (r = 0.744, df = 7, p = 0.021), indicating that fragmentation is constant over time (Jones, 2013; Lyman, 2008). Ecological indices of taxonomic diversity and evenness vary over time, suggesting complex variations in forager behaviour over time (Table 8). Generally, these indices have low values, indicating both low diversity and the dominance of a small number of taxa in the assemblage. This is largely controlled by the abundance of Neoradina prasongi, which dominate the assemblage in the lower levels, despite a greater number of other taxa also present. In the upper levels where Neoradina prasongi is absent, the diversity and evenness indices increase, but overall counts are low suggesting the site was less frequently used for subsistence activities.

Discussion

Geoarchaeology

The general picture indicated by the geoarchaeoloical data is one of subtle, uncoordinated changes in the variables we measured. This could be interpreted as indicative of either relatively constant conditions of deposition, or resulting from a high degree of bioturbation that has averaged out any major changes in the deposit. The sediment texture suggests a mixture of aeolian, colluvial and fluvial inputs, typical of cave and rockshelter deposits in the tropics (Westaway et al., 2009). The composition of the sediments varies little over time, as indicated by the measurements of organic matter, carbonates and pH in the bulk samples, and the ICP-AES data.

The XRD data show variation in the proportion of Kaolinite throughout the deposit. The Kaolinite is probably derived from the weathering of feldspars and other silicate minerals, and may relate to changes in weathering on the landscape around the site. Substantial changes in surface geochemistry are unlikely, due to the uncoordinated changes in magnetic susceptibility, which, if coordinated, might suggest episodes of soil formation on the landscape surrounding the site. The Carbon isotope values indicate a consistent dominance of C3 plants in the site environment through time, similar to the present-day environment. The small monotonic decrease in Carbon isotope values towards the present suggests that the deposit has some stratigraphic integrity, despite the anomalously low finds of ceramics. Thus, we can credibly interpret the geoarchaeological data as indicating generally constant conditions over time, rather than resulting from massive large scale bioturbation.

Faunal assemblage

KTC rockshelter has a relatively undisturbed record of mammalian, reptilian, piscean, and molluscan assemblages present in relatively intact deposits. The invertebrate record at the rockshelter provides a valuable description of molluscan taxa and subsistence patterns during the ‘missing millennia’. Neoradina prasongi molluscs were abundant during the late-Pleistocene, but much less frequent during the mid and late Holocene.

Neoradina prasongi molluscs live in fresh-water stream habitats (Brandt, 1974) . At KTC, N. prasongi shells constitute the bulk of molluscan food waste in the archaeological assemblage. The period of peak discard rates for N. prasongi is c. 9,000 cal BP, suggesting that the most intensive use of the rockshelter for subsistence purposes occurred during this time. Utilization of this mollusk also indicates that major freshwater stream habitats were near the rockshelter during the late-Pleistocene and early Holocene. Following their analysis of the faunal material from Lang Rongrien, Mudar and Anderson (2007) suggested that during the late-Pleistocene a drier and more open environment occurred in the Krabi region, this period was also characterized by increased monsoon seasonality. During the peak mollusk discard period of the late-Pleistocene at KTC, a drier and more open environment would have allowed forager groups to pursue large artiodactyls in the grassland-savanna habitat, alongside N. prasongi in seasonally abundant fresh-water stream resources. Occurrence of abundant turtle or tortoise remains at KTC also suggests that fresh-water stream habitats were found near the site throughout the late-Pleistocene and early-Holocene.

At Lang Rongrien, the abundant Testudines elements through time provided evidence of Southeast Asian turtle and tortoise exploitation occurred at the rockshelter (Mudar and Anderson, 2007). This record also appears at KTC with the presence of abundant Testudines specimens throughout the late-Pleistocene and Holocene deposits. However, this pattern has not been found elsewhere near KTC in southern peninsular Thailand. For example, both Moh Khiew Cave and Sakai Cave had very low abundances of Testudines remains during the late-Pleistocene and Holocene (Pookajorn, 1996). Generally, Testudines elements were only identified by presence or absence at these sites. Since Lang Rongrien and KTC both provide evidence of turtle and tortoise exploitation, this represents a supports the recognition of fresh-water stream foraging pattern. It also suggests that fresh-water stream habitats were abundant during the late-Pleistocene and early Holocene, in the Krabi region. Furthermore, previously identified shifts in monsoon seasonality, and intensity, during this time suggests that precipitation may have increased during the early Holocene in Peninsular Thailand (Marwick and Gagan, 2011), matching the faunal record at KTC.

The declining exploitation of freshwater N. prasongi molluscs into the Holocene, reaching a minimum at 6000 cal BP may reflect the shift from freshwater to mangrove swamp habitats in this region, or a shift in the foraging dynamics of prehistoric groups (Shoocongdej, 2000; 2010). The timing of the lowest amount of shells in the deposit coincides with the peak sea levels observed by Sinsakul (1992) and Tjia (1996). Rising sea-levels throughout the Holocene would have shifted mangrove environments closer to the rockshelter over time, which may have influenced the abundance and distribution of locally available resources and freshwater stream environments (Anderson, 1990; Horten et al., 2005; Tjia, 1996; Sinsakul, 1992). These initial faunal data from KTC describe a pattern of forager groups utilizing a diverse range of locally available taxa in the tropical rainforest environment, suggesting that foragers at KTC were able to effectively adapt to shifts in local environmental conditions. Additionally, our radiocarbon dates suggest that the decline in intensive harvesting of N. prasongi during the middle-Holocene, may be associated with the emergence of rice agriculture and farming in mainland Southeast appears (Castillo, 2011; Fuller, 2011; White et al., 2004). Thus, declines in mollusk utilization may reflect a pattern of rising sea-levels. The mechanism here may be a reduction in the availability of suitable mollusk procurement locations, favoring the adoption of agriculture during the mid and late-Holocene in peninsular Thailand as a response to these sea level changes. Shell exploitation picks up again at KTC at c. 3000 cal BP, coincident with the regressive phase at 3700 to 2700 BP described by Sinsakul (1992). This is also when site use changes, with more frequent visits suggested by peaks in the discard of archaeological materials.

Our data from KTC not only suggests that a subsistence change occurred at the Pleistocene-Holocene transition, but that foragers utilizing the rockshelter displayed a pattern of faunal exploitation not widely noted at archaeological sites in Thailand. Elsewhere in Thailand, large abundances of shellfish in rockshelter sites tend to date to the middle Holocene when a transition towards a broad-spectrum diet occurs, not during the terminal Pleistocene (Bulbeck, 2003). The earlier peak in the molluscan assemblage at KTC suggests that a different pattern of shellfish exploitation occurred here, one that we link to local environmental conditions controlled by sea level changes.

A broader implication of these results is that the patterns at KTC may offer some support to the model proposed by Hunt and Rabett (2014) for the transition from foraging to farming. They consider widespread forest disturbance in the Early Holocene as part of a trajectory toward predominantly agricultural subsistence. Using evidence from Borneo, they propose that evidence of palynological signatures of disrupted forest successions was linked to human translocation and propagation of economically-useful plants. Unfortunately our pollen and phytolith analysis was not informative about forest disturbance at KTC. However, the decline in the use of the site for exploiting mollusks may be part of a shift towards a greater focus on plant foods. We might speculate that as shellfish became less important in the diet of the KTC’s occupants, their pursuit of alternative resources initiated a distinct trajectory of economic change (Rabett, 2012). This may have involved a protracted process of wild plant food production (Fuller et al., 2007; Harris, 1989) or cultivation without domestication (Zhao, 2011), eventually resulting in reliance on farmed crops seen at Late Holocene sites in the region.

Conclusion

Excavations revealed human occupation at KTC from recent times back to over 13,000 years ago, without any major interruptions, disturbances or discontiunities. The changes in artefact technology were subtle during the time represented by the excavated deposit, and there is some uncertaintly about the effect of bioturbation on artefact distributions. That said, the site is unique because it has not been extensively disturbed by late Holocene human burials. The faunal assemblage proved the most abundant and interesting aspect of the excavated materials, and broadly confirms some of the patterns previously observed at Lang Rongrien. The most striking find is the correlation between the abundance of shellfish and past sea levels. Low sea levels at the early Holocene correspond to a peak in shellfish discard, followed by a decline in shellfish and lithic discard at c. 6000 cal BP, at the same time as the peak Holocene sea levels. There is another small peak in shellfish at c. 3000 BP during a regressive phase, this time accompanied by relatively large amounts of ceramics and lithics. This faunal discard sequence suggests that local sea levels influenced the intensity of site use. Past human occupants appeared to have found the site favorable for habitation during conditions of low sea levels. Presumably during higher sea levels they sought shelter further inland. In any case, we have shown that adaptation to sea level changes did not require major technological reorganization for the occupants at KTC, but instead was managed by adjusting settlement and land-use patterns to maintain access to resources such as shellfish.

The results from KTC confirm the ‘missing millennia’ as a period of important subsistence and technological changes in mainland Southeast Asia. One one hand, we see at KTC a recapitulation of a common sequence in mainland Southeast Asian prehistory. This includes foragers using the site for brief subsistence-related tasks during the late Pleistocene and early Holocene, then a transition in the middle Holocene to people using the site less for foraging activities, but now with ceramics and possibly practicing agriculture, as suggested by the polished adzes. On the other hand, we see a unique pattern of shellfish exploitation at KTC that is related to the local sea level changes. This relationship highlights the importance of local contingencies in understanding the mechanisms of change from foragers to agriculturalists. The model proposed by Hunt and Rabett (2014), of a locally contingent protracted process of human modification of plant resources, may be relevant in understanding how Early Holocene foragers at KTC relate to the Late Holocene occupants here, and potentially also sites such as Ban Non Wat. Future work on direct dating of the KTC ceramics will help to resolve questions the timing of their appearance here, and their relationship to shifts in subsistence behaviours.

Acknowledgments

Thanks to Boonyarit Chaisuwan (Fine Arts Department of Thailand) and Chawalit Khaokhiew (Silpakorn University) for assisting with access to the site. Thanks to Borisut Boriphon, Jessica Butler, Praewchompoo Chunhaurai, Anna Hopkins, Rachel Vander Houwen, Fitriwati, Kate Lim, Supalak Mheetong, Pham Thanh Son, Kim Sreang Em, Kyaw Minn Htin, and Chonchanok Samrit for helping to excavate the site and catalogue the finds. Thanks to Rodrigo Solinis Casparius, Pat Goodwin, David Hunt, Julia Malakie, Heather McAuley, Sherri Middleton, Hanyu Song, and Joss Whittaker for their assistence with the geoarchaeological laboratory analysis. Funding was provided by an ACLS/Luce Foundation grant to Peter Lape (University of Washington) and an International Provost grant to BM from the University of Washington Office of the Provost.

References

Alam, A., Xie, S., Saha, D., & Chowdhury, S. (2008). Clay mineralogy of archaeological soil: an approach to paleoclimatic and environmental reconstruction of the archaeological sites of the Paharpur area, Badalgacchi upazila, Naogaon district, Bangladesh. Environmental Geology, 53(8), 1639-1650.

Anderson, D. D. (1990). Lang Rongrien rockshelter: A Pleistocene, early Holocene archaeological site from Krabi, southwestern Thailand University of Pennsylvania Museum of Archaeology and Anthropology, University Museum Monograph, 71. Pennsylvania.

Anderson, D. D. (1997). Cave archaeology in southeast Asia. Geoarchaeology, 12(6), 607-638.

Anderson, D. (2005). The use of caves in peninsular Thailand in the Late Pleistocene and early and middle Holocene. Asian Perspectives, 137-153.

Araujo, A. G., Feathers, J. K., Arroyo-Kalin, M., & Tizuka, M. M. (2008). Lapa das boleiras rockshelter: stratigraphy and formation processes at a paleoamerican site in Central Brazil. Journal of Archaeological Science, 35(12), 3186-3202.

Arroyo-Kalin, M., Neves, E. G., & Woods, W. I. (2009). Anthropogenic dark earths of the Central Amazon region: remarks on their evolution and polygenetic composition. In Amazonian Dark Earths: Wim Sombroek’s Vision. Springer Netherlands. pp. 99-125.

Auetrakulvit, P., Forestier, H., Khaokhiew, C., & Zeitoun, V. (2012). New Excavations at Moh Khiew Site, Southern Thailand. In Crossing Borders: Selected Papers from the 13th International Conference of the European Association of Southeast Asian Archaeologists. NUS Press, Singapore. pp. 60-73.

Balcerzak, M. (2002). Sample digestion methods for the determination of traces of precious metals by spectrometric techniques. Analytical sciences, 18(7), 737-750.

Brandt, R.A.M. 1974. The nonmarine aquatic Mollusca of Tbailand. Arch. Molluskenkunde. 105, 1-423.

Bulbeck, F. 2003. Hunter-gatherer occupation of the Malay Peninsula from the Ice Age to the Iron Age. In: J. Mercader (ed.), Under the canopy: The archaeology of tropical rain forests. Piscataway: Rutgers University Press.

Carter, M. R. (1993). Soil sampling and methods of analysis. Boca Raton: Lewis Publishers.

Castillo, C. (2011). Rice in Thailand: the archaeobotanical contribution. Rice, 4(3-4), 114-120.

Chi, Z., & Hung, H. C. (2010). The emergence of agriculture in southern China. Antiquity, 84(323), 11-25.

Chitkament, T. (2007). Lithic analysis of Moh Khiew rockshelter (Locality I) in Krabi river valley, Krabi province, southwestern Thailand. Unpublished MA thesis, Department of Prehistory in the National Museum of Natural History, Paris, France.

Cook, S. F. (1965). Studies on the Chemical Analysis of Archaeological Sites. University of California Press, Berkeley.

Costa, M. L. and D. C. Kern, 1999. Geochemical signatures of tropical soils with archaeological black earth in the Amazon, Brazil. Journal of Geochemical Exploration, 66, 369–385

DeNiro, M. J. (1987). Stable Isotopy and Archaeology. American Scientist, 75, 2.

Eidt, R. C. (1985). Theoretical and practical considerations in the analysis of anthrosols. In (ed) G. Rapp Jr., Archaeological Geology, Yale University Press, New Haven, pp. 155–190

Forestier, H., Sophady, H., Puaud, S., Celiberti, V., Frère, S., Zeitoun, V., … & Billault, L. (2015). The Hoabinhian from Laang Spean Cave in its stratigraphic, chronological, typo-technological and environmental context (Cambodia, Battambang province). Journal of Archaeological Science: Reports, 3, 194-206.

Francis, C. (2008). A Guide to the Mammals of Southeast Asia. Princeton: Princeton University Press.

Fu, X. G. 2002. The Dingsishan site and the prehistory of Guangxi, southern China. Bull. Indo-Pacific Prehist. Assoc. 22:63–72.

Fuller, D. Q. (2011). Pathways to Asian civilizations: Tracing the origins and spread of rice and rice cultures. Rice, 4(3-4), 78-92.

Fuller, D.Q., R.G. Allaby, C. Stevens (2007). Domestication as innovation: the entanglement of techniques, technology and chance in the domestication of cereal crops. World Archaeology, 42(1),. 13–28

Gale, S. J., & Hoare, P. G. (1991). Quaternary sediments: petrographic methods for the study of unlithified rocks. New York: Belhaven Press.

Gorman, C. (1972). Excavations at Spirit Cave, North Thailand: Some interim interpretations. Asian Perspectives, 13, 79-107.

Harris, D.R. (1989). An evolutionary continuum of people-plant interaction, In D.R. Harris, G.C. Hillman (Eds.), Foraging and Farming: the Evolution of Plant Exploitation, Routledge, London, pp. 11–26

Hartman, G. (2011). Reconstructing Mid-Pleistocene paleovegetation and paleoclimate in the Golan Heights using the d13C values of modern vegetation and soil organic carbon of paleosols. Journal of Human Evolution, 60(4), 452-463.

Haslett, J., & Parnell, A. (2008). A simple monotone process with application to radiocarbon‐dated depth chronologies. Journal of the Royal Statistical Society: Series C (Applied Statistics), 57(4), 399-418.

Higham, C. (2013). Hunter-gatherers in Southeast Asia: From prehistory to the present. Human biology, 85(1), 21-43.

Higham, C., & Kijngam, A. (Eds.). (2011). The Origins of the Civilization of Angkor, Volume 4: The Excavation of Ban Non Wat. Part II: the Neolithic Occupation (Vol. 4). Fine Arts Department of Thailand.

Higham, C. F. W., & Thosarat, R. (2004).The Excavation of Khok Phanom Di, Vol. 7: Summary and Conclusions. London: Society of Antiquaries of London.

Higham, C. , Guangmao, X., & Qiang, L. (2011). The prehistory of a Friction Zone: first farmers and hunters-gatherers in Southeast Asia. Antiquity, 85(328), 529-43.

Horten, B.P., P.L. Gibbard, G.M. Milne, R.J. Morley and C. Purintavaragul. (2005). Holocene sea levels and palaeoenvirorunents, Malay-Thai Peninsula, southeast Asia. The Holocene 15 (8): 1199-1213.

Hunt, C. O., & Rabett, R. J. (2014). Holocene landscape intervention and plant food production strategies in island and mainland Southeast Asia. Journal of Archaeological Science, 51, 22-33.

Jones, E. (2013). Subsistence Change Among the 17th-Century Diné? A Reanalysis of the Faunas from the Fruitland Data Recovery Project. Journal of Ethnobiology 33(1):148-166.

Knudson, K.J. , L. Frink, B.W. Hoffman, T.D. Price (2004). Chemical characterization of Arctic soils: activity area analysis in contemporary Yup’ik fish camps using ICP-AES Journal of Archaeological Science, 31, 443–456.

Lampert, C., Glover, I., Hedges, R., Heron, C., Higham, T., Stern, B., Thompson, G. (2003). Dating resin coating on pottery: The Spirit Cave early ceramic dates revised. Antiquity, 77(295), 126-133.

Lerch, T. Z., Nunan, N., Dignac, M.-F., Chenu, C., & Mariotti, A. (January 01, 2011). Variations in microbial isotopic fractionation during soil organic matter decomposition. Biogeochemistry, 106, 1, 5-21.

Lyman, R. Lee (2008). Quantitative Paleozoology. Cambridge, Cambridge University Press.

McGrath, R. J., Boyd, W. E., & Bush, R. T. (2008). The paleohydrological context of the Iron Age floodplain sites of the Mun River Valley, Northeast Thailand. Geoarchaeology, 23(1), 151-172.

Marwick, B. (2016). “Computational reproducibility in archaeological research: Basic principles and a case study of their implementation”. Journal of Archaeological Method and Theory. in press, doi:10.1007/s10816-015-9272-9

Marwick, B., & Gagan, M. K. (2011). Late Pleistocene monsoon variability in northwest Thailand: an oxygen isotope sequence from the bivalve Margaritanopsis laosensis excavated in Mae Hong Son province. Quaternary Science Reviews, 30(21), 3088-3098.

Matsumura, H., & Pookajorn, S. (2005). A morphometric analysis of the Late Pleistocene human skeleton from the Moh Khiew Cave in Thailand. HOMO-Journal of Comparative Human Biology, 56(2), 93-118.

Midwood, A. J., & Boutton, T. W. (1998). Soil Carbonate Decomposition by Acid Has Little Effect on 13C of Organic Matter. Soil Biology & Biochemistry, 30, 1301.

Middleton, W.D. (2004). Identifying chemical activity residues on prehistoric house floors: a methodology and rationale for multi-elemental characterization of a mild acid extract of anthropogenic sediments Archaeometry, 46, 47–65

Middleton, W.D. and T.D. Price (1996). Identification of activity areas by multi-element characterization of sediments from modern and archaeological house floors using inductively coupled plasma-atomic emission spectroscopy Journal of Archaeological Science, 23, 673–687

Misarti, N., Finney, B. P., & Maschner, H. (2011). Reconstructing site organization in the eastern Aleutian Islands, Alaska using multi-element chemical analysis of soils. Journal of Archaeological Science, 38(7), 1441-1455.

Mudar, K., & Anderson, D. (2007). New evidence for southeast asian pleistocene foraging economies: Faunal remains from the early levels of Lang Rongrien rockshelter, Krabi, Thailand. Asian Perspectives, 46(2), 298-334.

Parnell, A. C., Haslett, J., Allen, J. R., Buck, C. E., & Huntley, B. (2008). A flexible approach to assessing synchroneity of past events using Bayesian reconstructions of sedimentation history. Quaternary Science Reviews, 27(19), 1872-1885.

Pookajorn, S. (1996). Human activities and environmental changes during the late Pleistocene to middle Holocene in southern Thailand and Southeast Asia. In: L. Guy Straus et al. (eds.), Humans at the End of the Ice Age. New York,

Pookajorn, S. (1994). Final report of excavations at Moh Khiew cave, Krabi province, Sakai cave, Trang province and ethnoarchaeological research of hunter-gatherer group so-called “Sakai or Semang” at Trang province. Bangkok, Department of Archaeology, Silpakorn University Press, Bangkok

Rabett, R. J. (2012). Human adaptation in the Asian Palaeolithic: hominin dispersal and behaviour during the Late Quaternary. Cambridge University Press.

Rabett, R., Appleby, J., Blyth, A., Farr, L., Gallou, A., Griffiths, T., & Tâń, N. C. (2011). Inland shell midden site-formation: Investigation into a late Pleistocene to early Holocene midden from Tràng An, Northern Vietnam. Quaternary International, 239(1), 153-169.

Reimer, P. J., Bard, E., Bayliss, A., Beck, J. W., Blackwell, P. G., Bronk Ramsey, C., & Grootes, P. M. (2013). IntCal13 and Marine13 radiocarbon age calibration curves 0-50,000 years cal BP. Radiocarbon 55(4), 1869-1887.

Sawada, J., Thuy, N. K., & Tuan, N. A. (2011). Faunal Remains at Man Bac, in Oxenham, M. F., & Matsumura, H. (eds) Man Bac: the Excavation of a Neolithic Site in Northern Vietnam. Terra Australis, 33, Australian National University E Press. pp. 105-116.

Schweizer, M., Fear, J., & Cadisch, G. (J1999). Isotopic (13C) Fractionation During Plant Residue Decomposition and its Implications for Soil Organic Matter Studies. Rapid Communications in Mass Spectrometry 13(13), 1284-1290.

Scott-Jackson, J. E., & Walkington, H. (2005). Methodological issues raised by laser particle size analysis of deposits mapped as Clay-with-flints from the Palaeolithic site of Dickett’s Field, Yarnhams Farm, Hampshire, UK. Journal of Archaeological Science, 32(7), 969-980.

Shoocongdej, R. 2010. Subsistence-Settlement Organization during the Late Pleistocene- Early Holocene: The Case of Lang Kamnang Cave, Western Thailand. In: B. Bellina et al. (eds.), 50 Years of Southeast Asian Archaeology: Essays in Honor of Ian Glover. River Books, Bangkok, pp. 51-66.

Shoocongdej, R. (2006). Late Pleistocene activities at the Tham Lod Rockshelter in Highland Pang Mapha, Mae Hong Son Province, Northwestern Thailand. In Uncovering Southeast Asia’s Past: Selected Papers from the 10th Annual Conference of the European Association of Southeast Asian Archaeologists. NUS Press, Singapore. pp. 22-37

Shoocongdej, R. (2000). Forager mobility organization ill seasonal tropical environments of western Thailand. World Archaeology. 32(1): 14-40.

Sinsakul, S. (1992). Evidence of quarternary sea level changes in the coastal areas of Thailand: a review. Journal of Southeast Asian Earth Sciences, 7(1), 23-37.

Sophady, H., Forestier, H., Zeitoun, V., Puaud, S., Frère, S., Celiberti, V., & Billault, L. (2015). Laang Spean cave (Battambang province): A tale of occupation in Cambodia from the Late Upper Pleistocene to Holocene. Quaternary International, doi:10.1016/j.quaint.2015.07.049 in press

Srisuchat, Amara (1997). Report of archaeological excavations in response to environmental impact of the construction of the hydroelectric project Cheo Lan Dam. Bangkok: Division of Fine Arts, Archaeology Division. (In Thai.)

Srisuchat, Tharapong and Amara Srisuchat (1992). Archaeological Analysis No.1: An Application of Technology and Science in Archaeological Work in Thailand. Bangkok: Division of Fine Arts, Archaeology Division.

Tieszen, L. L. (1991). Natural Variations in the Carbon Isotope Values of Plants: Implications for Archaeology, Ecology, and Paleoecology. Journal of Archaeological Science, 18, 3, 227

Tjia, H. D. (1996). Sea-level changes in the tectonically stable Malay-Thai Peninsula. Quaternary International, 31, 95-101.

Van Vlack, H. (2014). Forager subsistence regimes in the Thai-Malay peninsula: An environmental archaeological case study of Khao Toh Chong rockshelter, Krabi, Thailand. Unpublished MA thesis, Department of Anthropology, San José State University

Viet, N. (2007). The Da But Culture: evidence for cultural development in Vietnam during the middle Holocene. Bulletin of the Indo-Pacific Prehistory Association, 25, 89-94.

White, J. C. (2011). Emergence of cultural diversity in Mainland Southeast Asia: a view from prehistory. (ed) N. J. Enfield, Dynamics of human diversity. Canberra: Pacific Linguistics. pp. 9-46

White, J. C., & Bouasisengpaseuth, B. (2008). Archaeology of the Middle Mekong: introduction to the Luang Prabang province exploratory survey. Recherches Nouvelles Sur Le Laos. Vientiane and Paris, Études thématiques (18), 36-52.

White, J. C., Penny, D., Kealhofer, L., & Maloney, B. (2004). Vegetation changes from the late Pleistocene through the Holocene from three areas of archaeological significance in Thailand. Quaternary International, 113(1), 111-132.

Wickham, Hadley (2015). R Packages. Sebastopol, CA: O’Reilly Media, Inc.

Woods, W.I. (1984). Soil chemical investigations in Illinois archaeology: two example studies, in J.B. Lambert (Ed.), Archaeological Chemistry III, American Chemical Society, Washington, DC, pp. 67–77.

Woods, W.I. and B. Glaser (2004). Towards an understanding of Amazonian Dark earths, in B. Glaser, W.I. Woods (Eds.), Amazonian Dark Earths: Explorations in Space and Time, Springer, Berlin; London (2004), pp. 1–8

Wynn, J. G. (2007). Carbon isotope fractionation during decomposition of organic matter in soils and paleosols: Implications for paleoecological interpretations of paleosols. Palaeogeography, Palaeoclimatology, Palaeoecology, 251, 437-448.

Yoneyama, T., Okada, H., & Ando, S. (2010). Seasonal variations in natural 13C abundances in C3 and C4 plants collected in Thailand and the Philippines. Soil Science and Plant Nutrition, 56(3), 422-426.

Zhao, Z. (2011). New archaeobotanic data for the study of the origins of agriculture in China. Current Anthropology 52 (S4), S295–S306